The present invention relates to a method of forming a joint at an interface between two sintered bodies comprising multicomponent metallic oxides of specific crystal structure. When employing such sintered bodies in a device such as an oxygen separation device, it is often mandatory to join the same securely or even to provide a gas-tight joint, said joint being required to withstand operation conditions of the device. Typical sintered bodies in the above devices are an ion transport membrane (an electrolyte), an interconnect, a support, ceramic tubes, seals and conduits, etc. Such sintered bodies are typically joined tube-to-tube, tube-to-flat-plate and flat-plate to flat-plate, respectively.
Any joint is likely to form the weakest point of the entire device. Weak points are critical in cases where the device is subjected to severe operation conditions such as high temperature, high pressure differences or highly oxidizing or reducing environments which are tolerated by the sintered bodies themselves. To provide a commercially viable device, the joint is thus likewise required to maintain mechanical integrity, compatibility with the sintered bodies-and gas-tightness even when subjected to the operating conditions. Accordingly an ideal joint would possess comparable chemical and mechanical properties as the materials to be joined, especially comparable thermal cycling stability.
Up to now, joints between sintered bodies have been formed by using metallic brazes, nanocrystalline oxides, oxide-metal eutectics, glasses and ceramic-glass composites. See, e.g., S. D. Peteves et al., “The reactive route to ceramic joining: fabrication, interfacial chemistry and joint properties”, Acta mater. Vol. 46, No. 7, (1998), pp. 2407–2414; Y. lino, “Partial transient liquid-phase metals layer technique of ceramic metal bonding”, J. of Mat. Sci. Lett. 10, (1991), pp.104–106; S. Serkowski, “Application of ceramic-metal eutectics for solid-state bonding between ceramics,” Int. Symp. Ceram. Mater. Compon. Engines, 4th (Roger Carlsson et al. eds.) (1992) pp. 348–355; M. Neuhauser et al. “Fugen von Technischen Keramiken mit Keramik-Grunfolien,” Ber. DGK, Vol. 72, No.1–2, (1995) pp. 17–20; D. Seifert et al. “Verbind poroser mit dichtgesinterter Al2O3-Keramik durch Fugen mit keramischen Folien,” Ber. DGK, Vol. 73 No.10 (1996) 585–589; and R. Chaim et al. “Joining of alumina ceramics using nanocrystalline tape cast interlayer,” J. of Materials Research, 15, (2000) pp. 1724–1728. Joining of sintered bodies using ceramic-metal eutectics has the disadvantage of requiring the use of a metal. Many metals oxidize in air at high temperatures and therefore require the use of special reducing atmospheres to prevent the formation of a metal oxide. The sintered bodies to be joined may not be stable in these reducing atmospheres, which would result in decomposition of the sintered bodies. Joining of sintered bodies using nanocrystalline interlayers has the disadvantage of requiring very high pressures that could damage the parts to be joined due to creep or even fracture.
The use of brazes, i.e., metallic materials, or glasses, i.e., solid solutions of multicomponent metallic oxides, has the disadvantage of leaving behind an interfacial phase of the joint material with properties differing from, and in most cases inferior to, those of the materials being joined. For example, brazes leave behind a ductile metal, which at elevated temperatures can creep, be incompatible with the surrounding ceramic materials, or oxidize. Similarly, glass joints may have significantly different thermal expansion coefficients compared with surrounding multicomponent metallic oxides having perovskitic or fluoritic structure, resulting in undesirable residual stresses following temperature changes. Glass joints will further soften and flow at temperatures above their respective glass transition temperature. Finally, glass joints can be chemically incompatible with a sintered body of perovskitic or fluoritic structure at elevated temperatures. In any case, due to the remaining material, the joint will inevitably be visually or microscopically detectable, its properties being determined by the material of the joint itself, not the bodies to be joined.
Another method of forming a joint is disclosed in B. H. Rabin, and G. A. Moore “Reaction processing and properties of SiC-to-SiC joints”, Material. Res. Soc. Symp. Proc. 314, (1993), 197–203, Material Research Society, Pittsburgh. In this document it is disclosed that SiC components can be joined by using a mixture of Si and C powders. The document is silent on joining oxides in general, and especially on joining of multicomponent metallic oxides having fluoritic or perovskitic structure.
D. Seifert et al. “Verbind poroser mit dichtgesinterter Al2O3-Keramik durch Fugen mit keramischen Folien,” Ber. DGK, Vol. 73 No. 10 (1996) 585–589, discloses a  method to join alumina ceramics using ceramic joining foils of alumina-titania-calcia-magnesia. Other joining foils of alumina-titania-calcia-magnesia-silica and alumina-titania-manganese oxide-iron oxide-silica are also described. The joining temperature was greater than 100° K lower than the sintering temperature of the alumina ceramics to be joined. These joining compositions formed a liquid phase upon heating to the joining temperature. After joining, the joint retained the composition of the joining foils and was compositionally different than the alumina bodies that were joined. This reference states that the joining compositions to be used are highly specific to the ceramics to be joined. This reference is silent on how to join multicomponent metallic oxides. It is specifically silent on how to join perovskitic multicomponent oxides.
Another method to join alumina ceramics is disclosed in M. Neuhauser et al. “Fugen von Technischen Keramiken mit Keramik-Grunfolien,” Ber. DGK, Vol. 72, No.1–2, (1995) pp. 17–20. This method requires the use of ceramic foils made from a mixture of alumina, silica and other oxides. The presence of silica is undesirable since silica can be chemically or mechanically incompatible with the ceramics to be joined. In addition, this reference is also silent on how to join multicomponent metallic oxides.
A third method to join alumina parts using a (Al,Cr)2O3—Cr eutectic joining mixture is disclosed in S. Serkowski, “Application of ceramic-metal eutectics for solid-state bonding between ceramics,” Int. Symp. Ceram. Mater. Compon. Engines, 4th (Roger Carlsson et al. eds.) (1992) pp. 348–355. To obtain the joint, special gas atmospheres to produce extremely low oxygen partial pressures were required to allow the joining mixture to melt. The requirement of these special gas atmospheres limits the ceramics with which the eutectic mixtures can be used. Many ceramics will not be stable under the low oxygen partial pressure conditions needed for the eutectics to melt. Also the eutectic joining mixtures will result in the joint material being chemically and mechanically dissimilar to the bodies to be joined. This will have a negative effect of the stability and integrity of the joint. In addition, this reference is silent on joining multicomponent metallic oxides.
A fourth method to join alumina is disclosed in R. Chaim et al. “Joining of alumina ceramics using nanocrystalline tape cast interlayer,” J. of Materials Research, 15, (2000) pp. 1724–1728. This method requires hot pressing the alumina parts to be joined under uniaxial pressures of 55–80 MPa at 1200–1300° C. This method has the alleged advantage that the joint material is chemically and mechanically identical to the parts to be joined. However, the high pressures necessary to produce the joint are undesirable since the high pressures can lead to fracture or creep of the ceramic parts to be joined. In addition, this reference is also silent on how to join multicomponent metallic oxides.
In metallurgy, another type of bonding has been developed recently which is the so-called transient liquid phase bonding (TLP). See, e.g., Y. Zou et al., “Modelling of transient liquid phase bonding”, Int. Mat. Rev. Vol. 40, No. 5, (1995), p. 181, and I. Tuah-Poku et al., “Study of the Transient Liquid Phase Bonding, etc.”, Metallurgical Transactions A Vol. 19A, March 1988, p. 675. This process relies on the transient formation of a liquid phase depending on solute diffusion. The bonding technology has exclusively been used on metallic bodies.
It is therefore desired to provide a method of forming a joint between a first sintered body comprising a first multicomponent metallic oxide having a crystal structure of the perovskitic or fluoritic type and a second sintered body comprising a second multicomponent metallic oxide having a crystal structure of the same type as the first multicomponent metallic oxide, which method allows for formation of a joint that is chemically and mechanically compatible with the first and second sintered bodies. It is further desired that the formation of the joint does not leave behind a distinguishable interfacial phase. It is still further desired that the method should further allow for forming a compatible, refractory interfacial phase or joint, especially a joint exhibiting comparable thermal cycling stability.
It is also desired to provide a method of forming a joint between the above first and second sintered bodies, wherein the joint has similar chemical and mechanical properties as the sintered bodies to be joined or where the joint, if present in form of an additional phase, may even have a similar chemical composition and similar crystal structure as the first and second sintered bodies.
All references cited herein are incorporated herein by reference in their entireties.